13 Pull requests merged by 5 people

• adding rotary embedding example, with graph rewrite for complex subgraph

  #3570 merged Jul 3, 2025

• Lluo/cherry pick 3629

  #3656 merged Jul 3, 2025

• cherry pick 3642

  #3655 merged Jul 3, 2025

• feat: Refactor LLM model zoo and add KV cache support

  #3527 merged Jul 3, 2025

• Closed the perf gap of resnet and enabled refit

  #3629 merged Jul 2, 2025

• Fix dynamo core test failure on Windows

  #3642 merged Jul 2, 2025

• cherry pick 3636

  #3640 merged Jun 30, 2025

• release 2.8 branch cut

  #3638 merged Jun 30, 2025

• Temporary fix to workaround the mutable decomposition error.

  #3636 merged Jun 30, 2025

• upgrade torch_tensorrt version from 2.8.0.dev to 2.9.0.dev

  #3639 merged Jun 30, 2025

• fix the broken CC0 image link

  #3635 merged Jun 30, 2025

• disable aoti format in windows

  #3632 merged Jun 28, 2025

• disable test on aarch64 for now

  #3623 merged Jun 27, 2025

5 Pull requests opened by 4 people

• Removal of BAZEL build files from python package and changes to make cpp tests work [WIP]

  #3641 opened Jul 1, 2025

• feat: Add bf16 support to cast converter

  #3643 opened Jul 1, 2025

• Upgrade perf_run script to support TRT 10 and fix some issues

  #3650 opened Jul 2, 2025

• [kv_cache] integrated vlm code for benchmark (Stacked on #3527)

  #3652 opened Jul 3, 2025

• chore: Fix path in FLUX demo

  #3654 opened Jul 3, 2025

13 Issues closed by 2 people

• 🐛 [Bug] current BN implementation results in slower performance

  #3628 closed Jul 2, 2025

• 🐛 [Bug] Windows dynamo core tests failed

  #3631 closed Jul 2, 2025

• 🐛 [Bug] Test failure in tests/py/dynamo/conversion/test_abs_aten.py ModuleNotFoundError: No module named 'torch_tensorrt.dynamo.debug'

  #3575 closed Jul 1, 2025

• 🐛 [Bug] .tests.py.ts.api.test_classes

  #3576 closed Jul 1, 2025

• 🐛 [Bug] tests.py.dynamo.models.test_dtype_support.TestBF16Support.test_bf16_cpp

  #3579 closed Jul 1, 2025

• 🐛 [Bug] tests.py.dynamo.models.test_dyn_models.test_base_dynamic

  #3580 closed Jul 1, 2025

• 🐛 [Bug] FAILED automatic_plugin/test_automatic_plugin.py::TestAutomaticPlugin::test_mul_plugin_float_0 - AssertionError: Tensor-likes are not close!

  #3582 closed Jul 1, 2025

• 🐛 [Bug] FAILED runtime/test_002_lazy_engine_init.py::TestLazyEngineInit::test_lazy_engine_init_cpp_e2e - AssertionError: False is not true : Resnet18 TRT outputs don't match with the original model. Cosine sim score: nan Threshold: 0.99

  #3583 closed Jul 1, 2025

• 🐛 [Bug] FAILED runtime/test_003_cross_compile_for_windows.py::TestCrossCompileSaveForWindows::test_dynamo_cross_compile_for_windows_multiple_output - Failed: unexpected exception raised: Windows cross compilation feature is not available

  #3584 closed Jul 1, 2025

• 🐛 [Bug] MutableModule Test Failures

  #3586 closed Jul 1, 2025

• 🐛 [Bug] Dynamo core test failed in linux

  #3633 closed Jul 1, 2025

• 🐛 [Bug] Windows dynamo export serde test failed

  #3630 closed Jun 28, 2025

• 🐛 [Bug] Test failure `tests.py.ts.api.test_classes`

  #3578 closed Jun 28, 2025

14 Issues opened by 7 people

• 🐛 [Bug] examples/dynamo/torch_compile_gpt2.py run failed

  #3653 opened Jul 3, 2025

• 🐛 [Bug] FLUX Accuracy Issue

  #3651 opened Jul 2, 2025

• 🐛 [Bug] Draw svg for flux model hangs

  #3649 opened Jul 1, 2025

• 🐛 [Bug] group_norm implementation may slows down SD vae

  #3648 opened Jul 1, 2025

• 🐛 [Bug] index and index_put don't support dynamic shapes

  #3647 opened Jul 1, 2025

• 🐛 [Bug] Flux does not produce the correct image

  #3646 opened Jul 1, 2025

• 🐛 [Bug] with torch_tensorrt.dynamo.Debugger on torch_tensorrt.dynamo.compile it failed due to no graphviz installed

  #3645 opened Jul 1, 2025

• 🐛 [Bug] with torch_tensorrt.dynamo.Debugger is not working with MutableTorchTensorRTModule

  #3644 opened Jul 1, 2025

• ❓ [Question] Why is `torch.bfloat16` excluded from the `allowed_casts` set ?

  #3637 opened Jun 30, 2025

• 🐛 [Bug] Changing input size would affect the TRT engine size, testing on BERT

  #3634 opened Jun 28, 2025

• 📖 [Story] Performance improvements/gaps in Torch-TensorRT

  #3627 opened Jun 26, 2025

• 🐛 [Bug] Large Accuracy Issue

  #3626 opened Jun 26, 2025

• ✨[Feature] Integrate release pipeline with test pipeline

  #3625 opened Jun 26, 2025

• ✨[Feature] Automate examples verification in the release pipeline

  #3624 opened Jun 26, 2025

7 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• TRT-LLM loading mechanism tool

  #3398 commented on Jul 2, 2025 • 12 new comments

• ❓ [Question] Is SAM2 supported when compiling with the Dynamo backend on JetPack 6.1 or 6.2?

  #3478 commented on Jun 28, 2025 • 0 new comments

• error in data types

  #3614 commented on Jun 29, 2025 • 0 new comments

• 🐛 [Bug] Encountered bug when using Torch-TensorRT

  #3477 commented on Jun 30, 2025 • 0 new comments

• 🐛 [Bug] Weight streaming test failures

  #3585 commented on Jul 1, 2025 • 0 new comments

• Remove Bazel files from wheel

  #3615 commented on Jul 3, 2025 • 0 new comments

• Refit debug patch

  #3620 commented on Jun 27, 2025 • 0 new comments